Investigation of HIV-1 in the genital tract of women is necessary to develop effective vaccines, therapies, and strategies to block heterosexual and mother-to-child transmission. The rapid replication and evolution of HIV-1 coupled with selective pressures exerted by the host?s immune system leads to a genetically heterogeneous HIV-4 population known as a quasispecies. Quasispecies exist even in a single individual, with the degree of genomic variation between viruses from different individuals usually exceeding that seen within the same person. Such variation presents a major challenge to vaccine design. Preliminary studies of HIV-1 variation in the female genital tract vs. plasma provide evidence for viral compartmentalization, chronic dual infection with different strains in different compartments, and differential drug resistance. Investigation of the correlates of compartmentalization will help us to understand the mechanisms of containment and control of HIV-1 infection. To understand these separate HIV-1 reservoirs as well as mucosal transmission and immunologic control, it is crucial to investigate the interaction of HIV-1 and the immune response in both female genital tract and blood. We therefore propose to investigate the role of cytotoxic T cells (CTL) response in containing HIV-1 strains within the genital tract or blood. We plan to use state-of-the-art technology to focus on well-characterized women in the Women's Interagency HIV Study. We shall emphasize complementary, in-depth, virologic and immunologic studies of the same individuals. First, by analyzing contemporaneous HIV-I RNA sequences from genital tract and plasma, we shall identify women displaying HlV-1 compartmentalization, dual infection, and differential drug resistance. Then, we will investigate CTL responses in a subset of these women by examining the ability of each woman's systemic and genital tract CTL to recognize epitope variants representing the viral quasispecies present in her own two body compartments. We also propose to test whether differential recognition by CTL at these two sites is mirrored by differences in the epitope specificity of blood CTL with or without mucosa homing receptors. These studies promise to increase our understanding of the interactions between host CTL and virus, which are critical in controlling HIV-1 infection.